debug

data/models/llama3-1-70b.py

@@ -46,7 +46,9 @@ def chat_completion():
         
         outputs = model.generate(**inputs, do_sample=True, max_new_tokens=max_tokens,temperature=temperature,top_p=top_p,token=api_key).to("cuda")
 
-        return jsonify({"status": "success", "output": tokenizer.batch_decode(outputs[:, inputs['input_ids'].shape[1]:], skip_special_tokens=True)[0]})
+        response = tokenizer.batch_decode(outputs[:, inputs['input_ids'].shape[1]:], skip_special_tokens=True)
+
+        return jsonify({"status": "success", "output": response[0]})
     except Exception as e:
         return jsonify({"status": "error", "message": str(e)})